Comparison
Obsidian makes you link notes by hand. Memo reads everything you upload and draws the connections automatically — a living knowledge graph with semantic search, MCP, CLI and API.
The difference
With Obsidian
Obsidian is a brilliant local-first editor. But the graph is something you build by hand — every [[wiki link]], every backlink, every tag is manual work. Your PDFs and exports just sit there as attachments; nothing reads them for you.
With Memo
Upload a document and Memo reads it, extracts the people, organisations and concepts, and connects them to everything else you’ve filed. The graph builds itself, and the same entity across many documents becomes one node — automatically.
| Memo | Obsidian | |
|---|---|---|
| Building the graph | Automatic — extracted from your documents | Manual — you write the links |
| Reads PDFs & files | Yes — chunked, embedded, entity-extracted | Stored as attachments |
| Search | Semantic — by meaning, not just keywords | Full-text + plugins |
| Ask questions | Yes — cited answers across your corpus | Via third-party plugins |
| AI access | Built-in MCP, API and CLI | Community plugins |
| Setup | Sign in and upload — nothing to host | Local vault, sync add-on |
| Canonical entities | Resolved across documents | Per-note, manual |
Obsidian is a trademark of its respective owner. This comparison reflects default product behaviour and is provided for evaluation.
Why people switch
Stop maintaining a graph by hand. Memo extracts entities and relationships from each document and wires them together for you.
PDFs, web pages, notes and text are read, not just stored. Everything becomes searchable by meaning.
Ask a question in plain language and get an answer with citations drawn from your own documents.
Connect Claude and other MCP clients straight to your graph, or hit the HTTP API from your own tools.
Sign in with Google and upload. No vault to sync, no server to run — your graph lives in the cloud, privately.
Publish a read-only graph at a private link or embed it anywhere. Sharing is off by default and revocable.
Every upload runs through the same pipeline: Memo converts the document to text and splits it into passages, embeds each passage as a vector, extracts entities and typed relations with a language model, then resolves those entities to canonical nodes so the same concept across many files collapses into one. Read more in the knowledge graph guide, or see getting started.
Upload your first document and watch a knowledge graph build itself. Free to start; Pro when your library grows.